June 17, 2005
Early in
my GIS career, I
wanted to produce digital
maps at home.
Of course, I couldn't use tools I had access to at work due to
licensing and cost restrictions.This led me to investigate open
source GIS and mapping alternatives.I dove in and never looked back.
In fact, what began as a hobby has become a strategic
resource in
my consultancy work.I am bringing open source tools into the
corporate enterprise to replace or augment proprietary tools.It is
rather ironic that proprietary licensing pushed me to find
alternatives and that these alternatives are, in turn, supplanting
the proprietary.
The benefits of open source are a stark contrast to
closed source
options, particularly in the high-priced world of GIS software.The
following is a general overview of some positive benefits of open
source software.
What Is Open Source GIS?
Open source software refers to software projects that
have the
programming source code openly available for others to use.It is
free to use and customise, and is generally unrestricted in its
application.
Think of your favourite GIS and mapping programs for
desktop
analysis, web-based mapping, command line data conversion and
geospatial application programming.Then consider what it would be
like to have complete freedom to distribute and customise those
tools, with no licensing fees.Would it be encouraging, profitable
and maybe even fun? For me it was all three and more.Open source
software reminded me of why I studied GIS in the first place: to
investigate the world around me, for business or for pleasure.
Open source GIS and mapping projects are akin to
thousands of
other open source software projects, such as those hosted by
SourceForge.The principles they hold to
are summed up well by the Open
Source Initiative.Their site provides in-depth articles and
discussions about open source software.
Clearing the Open Air
The word "open," when referring to software, is
sometimes misunderstood and can mean different things to different
people.In software development there are two primary distinctions,
one being open source software, and the other being open standards or
protocols, designed to enhance interoperability.The term "Open
GIS"
usually refers to the work of the Open
Geospatial Consortium (OGC).
The OGC provides specifications for
sharing geospatial data and connecting applications, and has but only a
loose connection with open source software.In particular, OGC's
specifications are freely available, just as open source programs are,
and some open source projects implement those specifications.
If you are confused about the difference between
open
standards
and open source software, you are not alone.In some cases it is
particularly vague, implying the benefits of "open source"
when it is merely a reference to being able to share information with
others.
Why Consider Open Source Software
Using or developing open source software has five
advantages, which I list below.
This is not to suggest it is suitable for everyone.Your business
model, client base or other factors may not fit well with an open
source philosophy.A program you need may not be mature enough to use
in mission-critical applications, or your company might not have the
required internal expertise.
1.Community
Many projects have active communities.Support may come
from email
lists, live Internet Relay Chat (IRC) channels, commercial support
packages, books, etc.The most powerful aspect of the open source
community is that others also want the product to succeed.You have
instant allies for troubleshooting and sharing ideas.Volunteer
support is nothing new, even for proprietary products, but the
striking difference is that your voice is heard and appreciated.You
will probably even communicate with the programmer in charge of the
project.
When deficiencies are noticed, the discoverer is able
to help
define the problem or even fix it.Developers of the project have
publicly accessible bug tracking tools that users are encouraged to
populate. Like-minded users and developers are
able to work together,
sharing code or the costs of funding program improvements that
neither could afford individually. Open source
communities are active, encouraging, a great source
for finding new ideas and often the first to see innovate
applications of technology being developed.
2.Innovation
Open source communities and projects encourage
innovation.New
ideas, needs and problems you think are important are probably
already on the minds of others.Together, you can better define needs
and suggest changes to the developers.
Innovation is key for competitive business.There are
innovative applications for technology using closed source products,
but only as far as the product itself can be stretched.With open
source projects, the underlying technology can be improved and
customised, giving you a more competitive edge.Customised software
is in high demand.This demand cannot be easily met by closed source
programs.
3.Freedom
Open source projects use a variety of licensing
methods.Some are
more restrictive than others, but all aim to provide greater freedom.
The most common licensing limitation involves incorporating code into
your own proprietary product.Some open source licenses require you
to make your changes publicly available.For average users, this will
never be a problem.
In all cases, the freedom extends to allowing you to
use as many
copies of the software, on as many computers, by as many people as
you want.Contrast this with the closed source model where every
possible way of limiting users is employed to produce revenue from
licensing fees based on the number of users, computers, CPU, amount
of use and so on.
The increased freedom for programmers is undeniable.
Rather than
developing a whole project from scratch, existing open source
products or libraries can be used.Previously developed code can be
incorporated, speeding up project development.A programmer may also
choose to share a product with others, and in turn receive help
maintaining it.
Freedom to deliver products in many different ways
makes open
source a strategic option for companies looking to keep their clients
happy.For example, in the open source world you could use a product
like MapServer to serve
up a mapping program to a client.You could also take that product
and install it on the client's own network.You could make it more of
a stand-alone product, running only on your client's desktop
computer.You could put it on a CD or flash memory card and pass it
around, etc.You get the idea.You have the freedom to choose how to
deliver a product.Clients appreciate the sense of ownership and
flexibility.
4.Enjoyment
Any time you combine community, innovation and freedom,
projects
tend to be much more interesting, even fun.Many open source users
really enjoy being involved in community-based projects.Single
person consultancies can find it an opportunity to work with other
programmers.For corporations, staff are able to interact with others
in the field who share the same needs and interests.Working with a
broader community than your local sphere can be very satisfying.
Some believe that the more expensive a product is,
the better it is."You get what you pay for" doesn't always
hold true.Open source alleviates stress by reducing the cost.You can
then focus on the real tasks at hand:
implementation and customisation.
5.Cost
Costs for open source software are as good as it gets -
free.You
are not charged for using, improving or implementing open source
software.
Because the code is openly accessible, you can reduce
costs by
choosing to program internally rather than outsourcing.Likewise, if
you are not in a position to do modifications, you can hire expert
programmers who are familiar with the product, or others that you
know.
When you hire an open source developer, you are
guaranteed access
to the product long after that programmer is finished.Contrast this
with closed source applications.Move away from the parent company
and your investment is lost.With open source you have all the code
you paid to develop and can pass it on to another consultant in the
future.
Cost ties into all the other aspects mentioned in this
article,
especially innovation.Combining the benefits of licensing and costs,
you have an incredible opportunity to provide innovative solutions.
When costs and licensing are no longer a barrier, unlimited
possibilities are before you.
Other advantages
When source code is available, it can be ported to
other operating
systems and hardware platforms that a vendor might never support.
Longevity and stability are another positive aspect.By having
access to the source code you can choose what version to use and when
to upgrade.You can even take your version of choice and continue to
develop it on your own if desired.
There are many open source GIS and mapping products
available.
Below are a few good links to help you explore the options.
- OpenSourceGIS
software list
- FreeGIS data and
software list
- MapTools
|
Your Comments Post a comment All comments provided in this section are those of the individual who has created the post. These are not the opinions of Directions Media, its editors, staff or owners unless otherwise noted. Directions Media retains the right to edit or delete any comments posted herein.
|
|
||||||
| I'm afraid Mr. Baker propogates similar misconceptions on his part about open source software vs. commercial software. Sadly this is typical of comments from commercial vendors, and doesn't bring about neither common understanding or any consensus on the subject. Yes, open source software doesn't imply "free software" (as in "free beer"). This is an ages old discussion. There's normally a consultancy fee involved for building complex solutions with open source software. But that's as it should be, and the customer has a choice of doing it themselves or not. Even with commercial software there's more often than not consultancy involved. Using commercial software doesn't change that situation one bit. The main benefit of using open source software is not the cost of the software, but the wider economical margin for the consultant to build the right solution. By converting software cost to development cost it's possible to build a more customer specific solution. Usually that's what counts to the customer. There's no benefit in implementing a huge and complex software installation if you only need to employ less than a quarter of the embedded functionality. Commercial software tends to focus widely to facilitate a wider scope of off-the-shelf applications. Open source software tends to be more of a "bag of tools", that a skilled developer can turn into a specific solution. This distinction is the crux of why open source software is often successful over commercial software. Medium and smaller customers often need to provide a "custom" implementation rather than a standard one, and in that case open source software have an advantage. Larger customers almost never use open source software, since they often need and use the whole range of features in their implemented software, and commercial vendors often build customer specific solutions on this scale. Another huge benefit of open source development is precisely the lack of a singular software vendor. Anyone that has a need for a specific development can bring a bag of money to a skilled contractor, and have him do the work for them. I'm sure no major software vendor will perform customer specific changes in the base software for less than $5000. But that's possible with open source software. The need to always have some vendor to hold responsible is much over-rated, and constitutes pure scare tactics. Consultancy is always based on trust, and if someone breaks that trust, if you're using open source software, you can find.another consultant with precisely the same access to the base software as any other. Working with commercial software, one cannot just change the vendor this way, since they hold propriatary access to the software. And changing commercial software vendor is a very costly procedure indeed. And it certainly has nothing to do with shady characters trying to make themselves invaluable or procure themselves a huge profit. Quite frankly, such insinuations ought to be below any professional person's personal standard of conduct, and propogating such doesn't support your own cause, regardless of the validity of your other arguments. Commercial software still has a vital role to play in the market, namely by building platforms and reference implementations. Open source software doesn't need to participate on this level unless forced to do so. It may change over time, but that's how I see things are right now. My closing comment must be, that both types of software are here to stay, and everyone needs to find his own ground, and respect the other party. There are benefits for both sides to reap from employing the best solution in every case, and commercial software vendors can learn as much from open source developers/vendors as the other way around. |
||||||
|
||||||
| I really wished Mr. Baker took some time to research his arguments before posting a reply to an otherwise perfectly informative article. I also wished Mr. Baker didn't sound like he's defending the company he works for. Your company, Mr. Baker, sent two representatives to the just-concluded Open Source Geospatial Conference here in Minneapolis so it might have been more useful for you to have spoken to them before writing a "defensive" post. Mr. Tyler didn't write an article attacking proprietary software. Instead he wrote an article that provides an overview of the alternative to such software. If you feel offended by this article, you should at least respond with an equally informative set of arguments. Cheers! |
||||||
|
||||||
| Open Source can also provide significant advantages in developing countries. People can modify open source to suit local needs that might never be served by a commercial company. And if people in the developing countries are the ones who make the modifications, it results in an increase in the local expertise and potentially can have economic benefit to the community. Chris Holmes, Mick Wilson and I put together a paper on this which has some good references to other work on the same subject: http://www.eogeo.org/Projects/gsdi8paper-folder/holmes-doyle-wilson-gsdi-8.pdf |
||||||
|
||||||
| I would like to address some of Mr baker's points individually: "This article contains the typical, uncritical arguments promoting open source software. He does not mention any of the reasons to be cautious about open source software in GIS." Of course you should be cautious when deploying _any_ software in your organisation. Who wrote the software? What does the software do with my data? What information is leaving my premises via this software? Is the software suited to my useage requirements and so on. I dont believe proprietry software offers any more assurances than Open Source software does. In fact its very proprietry nature limits your ability to understand the true nature of the softare, and to make improvements to it. "1. No formal support. If you have a problem, there is no one you can hold responsible." Good informal support is often better than bad FORMAL support. How many times have you sat in a neverending hold queue waiting for your proprietry vendor to address your queries. How agile are they at responding to your queries? If you discover a bug, will they fix it and send you an update immediately, or will you have to wait eons while a multitude of bugs are aggregated into a nice money spinning 'upgrade' from your vendor? in short, I feel the 'support' issue is a card played too often by proprietry vendors who in many cases offer no real benifit over paid for support of Open Source software (and in most cases free support offered by the Open Source developer community) "2. Open source software in an area like GIS tends to be more complex and difficult to install than commercial software. That means more time and money invested than with commercial software." Taking two completely random examples, in my Debian system, I can say 'apt-get install grass' (or use a few mouse clicks in a gui package manager) and I have GRASS installed. On the other hand using a nameless software vendors license manager has in my experience been an excercise in pure frustration. I think Mr Baker's comment is completely anecdotal and not really based in the real world. |
||||||
|
||||||
| "3. Complex open source software is usually implemented by consultants. Those consultants don't work for free." No. This statement reveals numerous misconceptions. Firstly 'Free Software' relates to your freedom to view, modify and redistrubute the software, not to the price you pay for it (though by happy coincidence, most Open Source software is also free of charge!). Secondly, many many Open Source projects are indeed implemented by programmers who are working for free. This offers you a tangible benifit as the driving force behind the work of these programmers is enthusiasm and belief in the product, rather than a strong desire to make their monthly pay cheque. Thirdly there is nothing wrong with consultants working for a fee. What is very sad is when their effort is not made available to all other users of the software. What a waste of money, time and effort for two sites to pay for and develop the same software solution. "4. If you hire a consultant, you will likely have to rely on that consultant for maintenance and support--similar to commercial software. If you do the project in house, you're assuming that the talent implementing the project will be around when something breaks." Pure fallacy. Hiring a consultant is no more or less limiting than selecting a consultant who offers services based around a proprietry product. If you do development in house following the Open Source model, your developments would be contributed back to the software project, meaning that you can easily find a replacement developer who has intimate knowledge of the products codebase, and who can easily pick up where your in house developer left off. "5. Given projects of equivalent complexity, I think there is weak evidence that open-source implementations in GIS really save money over the long haul." One of the benifits of Open Source software is that its usually free of cost. So sure we dont have sales figures to back up our claims, but the very fact that we dont have sales figures speaks volumes to me about the amount of money people are NOT spending by using Open Source. "6. For the most part, open-source software has yet to produce software that exceeds commercial counterparts in terms of innovation. Mostly, open-source software duplicates the functionality of commercial software." While I'm not sure the substance of this claim has any merit, is innovation a prerequisite for choosing a software application? Where would Microsoft be today if no one bought Excel because it didnt offer great innovations over Lotus 1-2-3 ? Most people want good solid implementations of tried and tested solutions. Why use ArcIMS when you can use Mapserver? Mapserver is perhaps not innovative over ArcIMS, but its a good solid product that will cater to a large cross section of the market currently occupied by ArcIMS (and do it on considerably less powerfull hardware!). Even a cursory visit to a mapserver project like ka-maps will bear this claim out. "7. Programmers who contribute to open-source software make a living somehow. One should question any implication that it is unseemly to earn a living writing software." I dont think anyone has tried to imply that programmers are not entitled to earn a decent living. But customers are also entitled to pay for the true price of the software. not an arbitary price determined by the Marketing Dept. I discussed this more fully above. "If someone pitches open source software to you, ask yourself whether they have anything to gain from it. If they're a consultant, that should be obvious. If it's an in-house person, would the adoption mean some advantages for them or their group? Remember that something that sounds like a free lunch usually isn't." Well speaking from first hand experience as a person who contributes many hours of my spare time on QGIS, my only motive is because I'm keen to share the fruits of my labour and see people getting some use and enjoyment out of the software I participate in. My experiences with a great many other Open Software products have all been in a similar vein. Ive never had a developer try to 'hard sell' me some kind of paid for support for their Open Source product. Consultants offering services can and should quite reasonably paid for their time. Conversely how many people have been phone up by proprietry software salesperson trying to sell them software they dont need, cant afford and probably wont use? "Open source software has a place in the computing industry." Agreed. "However, we should question the idea that most or all software should be free." Why is the idea so questionable? Freedom to examine,extend and redistribute software should be a fundamental right of any user of that software. Rather we should question if there is really any valid case for proprietry software. How much further ahead would software be today if more of it was free, a snap shot of the contributions of all people who are interested in it? How much smaller would the first/third world divide be if people in the third world had equal access to computer software? It is really not that different from arguing that food, or clothing, or automobiles, or movies, or music should be free." Once again you are confusing 'free beer' with software freedom. Regards Tim Sutton |
||||||
|
||||||
| "7. ...If someone pitches open source software to you, ask yourself whether they have anything to gain from it. If they're a consultant, that should be obvious." If someone criticizes open-source software in an industry publication, ask yourself whether they have anything to gain from it. If they work for a proprietary software maker, that should be obvious. |
||||||
|
||||||
| The article by Tyler Mitchell, Open Source - Open Doors, is an interesting introduction to the open source, open systems software approach. Although ESRI is predominantly a closed source, open systems vendor we certainly acknowledge the role and interest in the open source movement, and appreciate the creative ideals of the open source community. |
||||||
|
||||||
| It would seem that Mr. Baker's comment, which is the basis for a number of other comments, is missing. So those who commented on his comment are kind of left hanging with dangling comments. Most of his comments have been preserved in the other comments, so at least they still more or less make sense. It's hard to motivate people to comment if they are given the impression that a site is a bit on the, shall we say, "ephemeral" side. |
||||||
|
||||||
| One of Baker's comments does sting a bit re the lack of innovation in open source. I realize now that while there is a lot of innovation in open source - it is hard to get marketing and presence required to make that innovation stick. I point you to Dave's spinny globe as an example of innovation for example: http://hm-geosphere.sourceforge.net/ - a |
||||||
|
||||||
| There's plenty of innovation in open source. Sticking just to the geo space, there's a range of stuff from GRASS to osgPlanet (OpenGL based spinning globes that use WMS to get data). There are little things like the CivicMaps javascript tiling engine, bigger things like MapBuilder, even bigger things like MapBender, uDig, MapServer, etc. These are not all "me too" projects. Many of them precede the commercial space and the ideas are assumed into the commercial space. Maybe people are not so familiar with these if their education consists of being taught how to use GIS packages from vendor X or Y (or if you're lucky from both X AND Y). Simply teaching how to use vendor packages could in fact be stifling innovation. If you want to get new ideas and new technologies, give the bright, young kids a bucket full of parts and see what they build. Don't give them a packaged system that prevents them from looking under the hood. |
||||||
|
||||||
| The post by Bryan Baker was removed at the request of Mr. Baker. It was not accidentally deleted and please do not repost his comments. Thank you. |
||||||
|
||||||
| Joe: No disrespect to you or Directions Magazine (and thanks for sponsoring the OSG '05 conference!) but Mr. Baker should have thought about the ramifications of his comments before he posted it. Deleting it takes away from the ensuing discussion. Fortunately, Tim Sutton quoted most of Mr. Baker's post. If anyone wants to see the original comment, google for "Open Source - Open Doors" and click on the cache. |
||||||
|
||||||
| I believe the community deserves a post from Mr. Baker clearly stating why he requested his comments be removed. His post was not graphic nor inappropriate nor was it bot-generated spam. Even though he keyed his employer in the "Company" line, he did state the comments did not reflect on his employer. Was he convinced that his points were poorly founded? Was he swayed by his employer because they may at some point in time deliver products which leverage Open Source (both free as in beer and free as in freedom) in their commercial offerings and thus these points undermine their credibility as well? Was he even aware his post was deleted at the request of somebody else? If he does maintain these beliefs he should repost the points without any reference to his employer. In an open forum, any hidden actions cause a never ending firestorm of conspiracy theories. Of course, in this case the proper editorial decision would have been to leave the comments but delete the company affiliation from the post. |
||||||
|
||||||
| I need a major project for B.E.(computer)for final year in java programming language. |
||||||
|
||||||
| Somebody has to have these comments copied somewhere? Can they be posted to another blog somewhere. Poor decision by Directions to retract without explanation. |
||||||
|
||||||
| As a program manager in SAIC with a program that entirely uses open source solutions for a Federal Government client, I thought it would be beneficial to follow-up on one of Lars' statements. Quote: "The main benefit of using open source software is not the cost of the software, but the wider economical margin for the consultant to build the right solution. By converting software cost to development cost it's possible to build a more customer specific solution. " This is very true AND I would add, when dealing with organizations that have a smaller budget, often given one must hire consultants to implement a solution for any large package/solution, ridding yourself of an expensive and proprietary license oftem makes dramatic sense. If you only have a $100,000 budget and a commercial solution package eats up 70% of that budget, but gets you 80% there; you now are left with only 30% of the money to make the customizaion. Additionally, once the solution is in place, you most likely will be paying 10-20% per year thereafter just for the right to run the solution, even if you never need change it. Open source software provides you with the ability to not only avoid the initial large license fee (which agreeably will most likely at least partially be consumed by additional consulting fees), you at least avoid the restrictions and future license fees. Additionally, you have the freedom to modify any part of the source code needed. One other point to make, by contributing your modifications back to the community, you can gain support from within the community, this helps you avoid being dependent on your original consultant that developed the solution and expands the talent base that can support it. Paul Boos Program Manager SAIC PS - These are my views alone; my company uses, supports, contributes to, and occasionally resells solutions that are commercial/proprietary in nature, are based on open source solutions, or contain an integrated mix of open source and proprietary solutions. |
||||||
